1. Field of the Invention
The present invention relates to a document management system, in which a document management server searches a document using index information.
2. Description of the Related Art
In a document management system for managing documents by associating the documents with index information, index search is generally used for searching the document including the index information corresponding to a keyword specified by a user. The index information includes a plurality of types of index definitions (also referred to as “index types”, “item names” or “attribute names”) and a value of each index definition (also referred to as an “index value” or “attribute value”) given to each document. The index definition (type, item name, and attribute name) generally defines the index type necessary for each document management unit (e.g., for each database or each management hierarchy) instead of using the same definition for the entire document management system (all database).
However, the management becomes too complicated if the index is freely defined for various kinds of hierarchy. Thus, in some systems, the hierarchy is provided in which the index can be defined, and the index can be defined for each management unit in the hierarchy. When the index is searched in such a document management system, the management unit is specified as a search target, the user specifies a search condition (index definition and search keyword), and then the search is performed based on the specified search condition.
Further, in Japanese Patent Application Laid-Open No. 2010-097296, in step S401, when specification of a search target area where a document file is searched is received, in steps S402 and 403, a file name, position information, and a plurality of attribute names and attribute values of the document file included in the specified search target area are acquired and stored.
When an input of a search word (search keyword) is received in step S404, then in step S405, a plurality of attribute values corresponding to the search word are extracted from among the stored attribute values. In step S406, a plurality of attribute names corresponding to the plurality of extracted attribute values are acquired, and in step S407, the acquired attribute names are displayed in a list. When, in step S408, the attribute name desired by the user is selected from the displayed list, in step S409, a document file including the selected attribute name and the search word is extracted.
In steps S402 and S403 of Japanese Patent Application Laid-Open No. 2010-097296, since the file name, the position information, and the plurality of attribute names and attribute values need to be acquired for all the document files included in the search target area, if the search target area is set in a wide area, processing for acquiring the information takes time.
Further, in the system in which the index type can be defined for each document management unit, when the plurality of document management units are set as the search target area, the processing for acquiring the information about the index definition (information about the index type) alone takes much time. For example, if the management units in which ten index types are defined include ten kinds of hierarchy, when acquiring the information about the index definition from each hierarchy (each management unit) takes 0.5 seconds, then acquiring index definition information from ten kinds of hierarchy takes 5 seconds.
Thus, in such a system where the user inputs the desired index definition and the search keyword after the information about the index definition is acquired, the user needs to wait a while until the system acquires the information about all the index definitions.